Clinical Studies:The role of cholinergic system in mood disorders has been highlighted recently through our demonstration that blocking cholinergic muscarinic activity with scopolamine produces rapid antidepressant effects. Our early studies identified an antidepressant effect of scopolamine in both MDD and BD patients. Significant (p< 0.001) clinical improvement occurred following a single infusion of scopolamine. Over the course of the clinical trial, 70% of patients experience a full response (50% reduction in symptoms) and over 50% experience remission of symptoms.A replication of our original finding was published in an independent MDD sample of patients. Moreover, we observed that while men and women show significant clinical improvement following scopolamine, women show a larger response than do men (p< 0.01). Recently, we conducted a pharmacokinetic study to identify an alternative route of scopolamine administration. The NIH pharmacy developed a nasal spray formulation, and over the past year we completed the collection of pharmacokinetic data to identify the nasal spray dose that most closely approximates the area-under-the-curve obtained with the i.v. administration. These samples currently are being processed by our collaborator, Irving Wainer, PhD. Clinical trials utilizing this route of administration potentially will begin within the next year. Cognitive and Imaging Studies: Behavioral and cognitive features of depression are associated primarily with the processing of affective information. A consistently reported finding is a mood congruent processing bias, which is defined as a tendency to show a bias for processing negative as compared to positive or neutral information. The mood congruent processing bias observed in MDD can be characterized within the framework of cholinergic system and stimulus processing mechanisms. The biased processing of negative or sad information is consistent with over-active cholinergic function in depression resulting in the over-representation of negative information. We would hypothesize that competition among stimuli in the environment engages cholinergic system, and the overactive system alters the bias preferentially towards negative stimuli in MDD. We are utilizing functional brain imaging methods to elucidate the role of the cholinergic system in stimulus processing biases. We characterized behaviorally a stimulus processing bias associated with a selective attention task where face and house stimuli are presented simultaneously. Faster reaction time when attending to faces (vs houses) reflects a stimulus process bias towards faces. Scopolamine selectively increased reaction time when attending to face stimuli with no change when attending to houses, reflecting a stimulus specific shift in the processing bias. The brain regions in visual processing areas which show a bias towards faces (larger response to faces than to houses) during placebo, reduce the bias during scopolamine, a result that complements the behavioral finding. In brain regions that show a bias for processing houses during placebo, no change is observed following scopolamine. Acetylcholine also influences working memory (WM) function through stimulus processing mechanisms and particularly is critical to the encoding of information through stimulus processing mechanisms. We evaluated the influence of blocking cholinergic muscarinic activity on neural responses during WM encoding as the attended stimulus feature was modulated in a fMRI study. During the task a picture of a face was shown (encoding) followed by a delay component (maintenance) then another picture of a face (test/retrieval). Healthy participants were instructed to match the test stimulus to either the identity or the emotional expression of the encoded face. Neural activity associated with encoding was estimated for the emotion and identity task conditions. Blocking cholinergic muscarinic reccceptors preferentially reduced BOLD response during encoding selectively when attending to the emotional stimulus content, with no change when attending to identity. This finding suggests that the more salient stimulus feature (i.e. emotion) is more susceptible to cholinergic modulation. Biomarker of Treatment Response Clinical Markers of Treatment Response. We evaluated the potential for baseline clinical ratings to predict treatment response to scopolamine using discriminant function analysis to determine if a linear combination of self-report mood-ratings could discriminate treatment responders from non-responders. The disciminant function identified a set of 4 or 5 clinical ratings that, when combined into a linear function, classified 86% of unipolar patients as responders or non-responders (p= 0.002); the discriminant function scores differed significantly between the responders and non-responders (p< 0.001). A validation analysis was able to correctly predict response classification in over 70% of MDD patients. Similarly, the discriminant function classified 88% of BP patients, and the discriminant fuction scores differed significantly between the responders and non-responders (p< 0.001). The validation analysis in the BP group was able to correctly predict classification in over 85% of the patients. We also are considering potential clinical or patient variables that might contribute to the identification of patients who will response to scopolamine treatment. Previously we demonstrated that gender influenced treatment outcome, and we currently are preparing a manuscript that shows that patients with treatment resistant depression show a comparable rapid clinical response to scopolamine as do treatment naive patients, although the treatment naive group continued to improve over additional infusions. Functional Imaging Markers of Treatment Response. Previously we identified areas of bilateral middle occipital cortex where baseline levels of cortical activity under specific task conditions correlated with subsequent treatment response to scopolamine in patients with MDD. Following treatment with scopolamine, changes in neural response in these same brain regions also correlated with the magnitude of response. Recently we used a selective attention task and emotion modulation to estimate emotional processing biases in the same cortical regions prior to treatment with scopolamine, and similarly observed correlations between neural response to specific task conditions and subsequent treatment outcome in patients with MDD. These findings indicate that difference in response patterns based on emotional content of information has the potential to predict subsequent response to treatment with scopolamine. Subsequently, we looked at baseline activity in the same brain regions during performance of an emotion evaluation task prior to treatment with ketamine, which is another rapid acting antidepressant agent. Again, we observed that the same brain regions that predicted treatment outcome to scopolamine also predicted treatment outcome to ketamine, in an independent cohort of patients with MDD and using a different emotion processing task. Together, these studies point to the potential for neural activity in middle occipital cortex during emotional processing to provide a biomarker of treatment response across rapid antidepressants.